Data engineer, Paris · Airflow · ClickHouse · dbt · Kubernetes · Snowflake · Databricks · Spark · AWS · GCPData engineer, Paris · Airflow · ClickHouse · dbt · Kubernetes · Snowflake · Databricks · Spark · AWS · GCP

Field notes on what breaks, what works, what we learn running these in production. → Follow me on LinkedIn

LATEST ARTICLES

See all articles →
  1. 2026-09-0376.6% for $52 or 65.7% for $1: What ClickHouse's AI Agent Benchmark Actually ShowsData Engineering · 7 min
  2. 2026-08-31What Should a Data + AI Architecture You'd Actually Recommend to a Company Look Like in 2026?Data Engineering · 10 min
  3. 2026-08-14Migrating Airflow to Kubernetes: The Real Challenge Wasn't TechnicalData Engineering · 5 min
  4. 2026-08-12Debugging Blocked ClickHouse DDL with ON CLUSTERClickHouse · 3 min
  5. 2026-08-12Data Pipeline Idempotence: The Night a Backfill Doubled Our RevenueData Engineering · 3 min

ABOUT

I'm Tahirintsoa Mamitiana. Data engineer in Paris, I design and run data pipelines and cloud architectures on AWS and GCP, with Airflow, ClickHouse, dbt, Kubernetes or Terraform depending on what the project needs. I grew up in Fianarantsoa, in the Malagasy highlands, before moving to France. This blog is a set of technical field notes, because I'd rather share what breaks and what works than a list of skills.